Nilimajain
commited on
Commit
•
6f1e734
1
Parent(s):
ab07df5
Nilimajain/gemma-ft-qa-newnj
Browse files
README.md
CHANGED
@@ -19,6 +19,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
19 |
# new_genna_nodel
|
20 |
|
21 |
This model is a fine-tuned version of [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it) on the generator dataset.
|
|
|
|
|
22 |
|
23 |
## Model description
|
24 |
|
@@ -44,10 +46,17 @@ The following hyperparameters were used during training:
|
|
44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
- lr_scheduler_type: constant
|
46 |
- lr_scheduler_warmup_ratio: 0.03
|
47 |
-
- num_epochs:
|
48 |
|
49 |
### Training results
|
50 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
51 |
|
52 |
|
53 |
### Framework versions
|
|
|
19 |
# new_genna_nodel
|
20 |
|
21 |
This model is a fine-tuned version of [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it) on the generator dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 2.7368
|
24 |
|
25 |
## Model description
|
26 |
|
|
|
46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
- lr_scheduler_type: constant
|
48 |
- lr_scheduler_warmup_ratio: 0.03
|
49 |
+
- num_epochs: 5
|
50 |
|
51 |
### Training results
|
52 |
|
53 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
54 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
55 |
+
| 2.335 | 1.0 | 649 | 2.4869 |
|
56 |
+
| 2.1888 | 2.0 | 1298 | 2.5154 |
|
57 |
+
| 2.0938 | 3.0 | 1947 | 2.5809 |
|
58 |
+
| 1.9877 | 4.0 | 2596 | 2.6546 |
|
59 |
+
| 1.8859 | 5.0 | 3245 | 2.7368 |
|
60 |
|
61 |
|
62 |
### Framework versions
|
adapter_config.json
CHANGED
@@ -20,12 +20,12 @@
|
|
20 |
"rank_pattern": {},
|
21 |
"revision": null,
|
22 |
"target_modules": [
|
23 |
-
"
|
24 |
-
"
|
25 |
"o_proj",
|
26 |
"q_proj",
|
27 |
-
"
|
28 |
-
"
|
29 |
],
|
30 |
"task_type": "CAUSAL_LM",
|
31 |
"use_dora": false,
|
|
|
20 |
"rank_pattern": {},
|
21 |
"revision": null,
|
22 |
"target_modules": [
|
23 |
+
"k_proj",
|
24 |
+
"gate_proj",
|
25 |
"o_proj",
|
26 |
"q_proj",
|
27 |
+
"v_proj",
|
28 |
+
"up_proj"
|
29 |
],
|
30 |
"task_type": "CAUSAL_LM",
|
31 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14331480
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:38dcedee9b883270d9ebc64e956b3e5465f913620072aa1fff8785d0d19936f5
|
3 |
size 14331480
|
runs/Aug18_05-09-36_0954fd5125d6/events.out.tfevents.1723957949.0954fd5125d6.1083.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d29cc05d520c375624f3475dcba269f60c43c13fba0c2e31566ec3cd0cdd1973
|
3 |
+
size 34676
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5432
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fddcdf5a1f078a8373cc45a03b990692d9c93559a6bc897cd196bed4997948ef
|
3 |
size 5432
|